1. Quality Control, Trimming & Preprocessing

FastQC evaluates raw sequencing quality across 11 diagnostic modules: per-base sequence quality, per-tile quality, per-sequence GC content, duplication levels, overrepresented sequences, and adapter contamination. Run in parallel across paired-end FASTQ archives.

# 1. Create QC output directory
mkdir -p fastqc_raw_reports/

# 2. Run FastQC in parallel on multiple paired-end FASTQs with custom adapters
fastqc \
  --threads 8 \
  --outdir fastqc_raw_reports/ \
  --extract \
  sample1_R1.fastq.gz sample1_R2.fastq.gz \
  sample2_R1.fastq.gz sample2_R2.fastq.gz

# 3. Quickly verify PASS / WARN / FAIL statuses across all samples in terminal
for summary in fastqc_raw_reports/*/summary.txt; do
    echo "=== Sample: $(dirname $summary) ==="
    awk '{printf "%-6s : %s\n", $1, $2}' "$summary"
done

MultiQC parses log outputs and summary statistics from dozens of independent bioinformatics tools (FastQC, fastp, Samtools, Cutadapt, Bowtie2, STAR) and synthesizes them into an interactive, publication-ready HTML dashboard.

# 1. Aggregate QC logs across all sample directories into an interactive dashboard
multiqc . \
  --outdir multiqc_dashboard/ \
  --filename cohort_qc_report.html \
  --title "Cohort NGS Quality Metrics" \
  --interactive \
  --export \
  --zip-data-dir

# 2. View summary metric table directly in terminal
head -n 25 multiqc_dashboard/cohort_qc_report_data/multiqc_general_stats.txt

Written in multi-threaded C++, fastp is the modern gold standard for short-read preprocessing. It performs adapter trimming (auto-detects paired-end adapters from overlap), quality trimming (sliding window), poly-G clipping (crucial for NovaSeq 2-color chemistry), UMI parsing, and outputs interactive HTML/JSON reports in a single pass.

# Production paired-end execution with auto-adapter detection and polyG tail clipping
fastp \
  --in1 sample_R1.fastq.gz \
  --in2 sample_R2.fastq.gz \
  --out1 clean_R1.fastq.gz \
  --out2 clean_R2.fastq.gz \
  --detect_adapter_for_pe \
  --trim_poly_g \
  --poly_g_min_len 10 \
  --cut_front \
  --cut_tail \
  --cut_window_size 4 \
  --cut_mean_quality 20 \
  --qualified_quality_phred 20 \
  --unqualified_percent_limit 30 \
  --length_required 50 \
  --html sample_fastp.html \
  --json sample_fastp.json \
  --thread 8

Trimmomatic provides granular step-by-step pipeline control for Illumina data. It leverages palindrome mode for reading through short fragments, clips low-quality read tails, and properly partitions paired reads from orphans whose mate failed quality filtering.

# Trimmomatic paired-end adapter clipping and sliding-window quality trimming
trimmomatic PE \
  -threads 8 \
  -phred33 \
  sample_R1.fastq.gz sample_R2.fastq.gz \
  clean_paired_R1.fastq.gz clean_unpaired_R1.fastq.gz \
  clean_paired_R2.fastq.gz clean_unpaired_R2.fastq.gz \
  ILLUMINACLIP:/usr/share/trimmomatic/TruSeq3-PE-2.fa:2:30:10:2:keepBothReads \
  LEADING:3 \
  TRAILING:3 \
  SLIDINGWINDOW:4:20 \
  MINLEN:36

Cutadapt specializes in finding and removing adapter sequences, primers, poly-A tails, and unique molecular identifiers (UMIs) with semi-global alignment and configurable error tolerance. Widely used in small RNA-seq, amplicon panels, and custom library constructs.

# 1. Paired-end Illumina TruSeq adapter trimming with error tolerance
cutadapt \
  -a AGATCGGAAGAGCACACGTCTGAACTCCAGTCA \
  -A AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT \
  -q 20,20 \
  --minimum-length 35 \
  --overlap 5 \
  --error-rate 0.1 \
  --cores 8 \
  -o clean_R1.fastq.gz \
  -p clean_R2.fastq.gz \
  raw_R1.fastq.gz raw_R2.fastq.gz \
  > cutadapt_trimming_report.txt

# 2. Trim 3' Poly-A tails in mRNA/Single-cell libraries (minimum 15 A's)
cutadapt \
  -a "A{15}" \
  --minimum-length 30 \
  --cores 4 \
  -o polyA_clean_R1.fastq.gz \
  raw_R1.fastq.gz

Designed specifically for long reads (Oxford Nanopore Technologies and PacBio HiFi/CLR). NanoPlot computes N50, total throughput, median read length, and generates bivariate length-vs-quality plots and pore activity heatmaps.

# 1. Profile Oxford Nanopore FASTQ reads with interactive hexbin plots
NanoPlot \
  --fastq ont_sample.fastq.gz \
  --outdir nanoplot_qc_report/ \
  --prefix ONT_Run_ \
  --plots hex dot \
  --loglength \
  --minlength 500 \
  --minqual 8 \
  --threads 8 \
  --format png html

# 2. Ultra-fast direct parsing of ONT sequencing_summary.txt (no FASTQ decompression needed)
NanoPlot \
  --summary sequencing_summary.txt \
  --outdir nanoplot_summary_report/ \
  --readtype 1D2 \
  --threads 8

Oxford Nanopore reads often contain residual sequencing adapters or chimeras. Porechop identifies and strips end adapters and splits chimeric reads with internal adapters. Chopper (written in Rust as a high-speed NanoFilt successor) streams FASTQ files for length and mean Phred quality filtering.

# 1. Demultiplex and strip ONT adapters using Porechop
porechop \
  -i raw_nanopore_reads.fastq.gz \
  -b ./demultiplexed_barcodes/ \
  --barcode_diff 5 \
  --threads 8 \
  --format fastq.gz

# 2. Streamline Chopper filtering (Rust): Min length 1000 bp, Max length 60,000 bp, Quality >= Q10
zcat ./demultiplexed_barcodes/BC01.fastq.gz | \
  chopper \
    --minlength 1000 \
    --maxlength 60000 \
    --quality 10 \
    --threads 8 | \
  gzip > BC01_clean_Q10.fastq.gz

FastK (Gene Myers) is an ultra-fast, memory-lean k-mer counter tailored specifically for both short (Illumina) and long (PacBio HiFi, Oxford Nanopore) raw reads without pre-correction. It counts canonical k-mers (up to k=50), estimates genome size and heterozygosity, and profiles sequencing error rates.

# 1. Count canonical 40-mers on raw sequencing reads (Illumina or PacBio HiFi)
FastK \
  -k40 \
  -t8 \
  -M32 \
  -bc1 \
  -p sample_reads.fastq.gz \
  -Nsample_k40

# 2. Generate k-mer frequency histogram (input for GenomeScope 2.0 genome size estimation)
Histex -G sample_k40 > sample_k40.hist

# 3. Query specific k-mer counts or extract top-frequency k-mer table
Tabex -t sample_k40 | head -n 30